Electronic Commerce and Developing...

THE INTERNET IS GLOBALIZATION ON STER-oids. It will boost efficiency and en-hance market integration domestically

and internationally, particularly in developingcountries that are most disadvantaged by pooraccess to information. Although the Internetshould enhance global growth, it also brings in-creased danger of economic marginalization tocountries that cannot access it effectively. Tak-ing advantage of electronic commerce requirespolicies similar to those needed to capitalize onthe opportunities for trade: improved interna-tional coordination, for example in ensuringinteroperability of communications technologyand confronting challenges to domestic tax andfinancial systems; an open economy promotingcompetition and diffusion of Internet technolo-gies; and efficient social and infrastructure ser-vices, in particular a competitive telecommuni-cations sector and a well-educated labor force.

Despite the obvious benefits of the Internet,uncertainty exists about the implications ofthis technology and its likely rate of diffusion.This chapter provides a tentative view of theimplications of electronic commerce for devel-oping countries, based on the theoretical liter-ature, inferences from experience in industrialcountries, and anecdotal evidence. The discus-sion is inevitably somewhat more speculativethan in other chapters. The evidence, however,warrants four broad conclusions:

Firms in developing countries should enjoyproductivity gains and expanded demand withthe spread of electronic commerce.

The Internet will boost productivity in de-veloping countries by increasing the efficiencyof the procurement system, strengthening in-ventory control, lowering retail transactioncosts, and eliminating or transforming inter-mediaries. Virtual proximity to industrialcountry markets will increase as the Internetreduces the costs inherent in operating at adistance. Given the wide differences in returnsto factors in developing versus industrialcountries, this increased proximity will gener-ate large gains from trade in sectors that lendthemselves to electronic commerce.

Consumers will benefit from increasedcompetition and market transparency, but thebenefits to firms will vary greatly, dependingon the sector, degree of product differentia-tion, and level of technological sophistication.

Developing country firms that sell labor-intensive, differentiated products (such ascrafts, software, or business services—particu-larly services involving the remote processingof routine information) will experience in-creased demand. These firms also will benefitfrom the opportunity to leapfrog to the mostadvanced technologies and from easier accessto advertising on global markets.

The impact of electronic commerce on devel-oping countries’ sales to global supply chains isuncertain. Reduced transactions costs shouldprovide greater interaction among multi-nationals and technologically sophisticatedfirms in developing countries. Many developing-country firms may lack the reputation required

1

Electronic Commerce andDeveloping Countries

4

Embargoed until Tuesday, December 5, 2 p.m. EST

Unpublished Proofs

to bid on the newly created online exchanges,however. Industrial-country multinationals alsomay prefer integrating their operations moreclosely with a reduced number of the most ad-vanced firms, given the opportunities for mana-ging tightly linked production processes throughthe Internet.

Government action is critical to removingimpediments to electronic commerce.

Network externalities imply that marketprices may not fully reflect the gains to the so-ciety from increased levels of Internet access;hence the government has a role in speedingInternet diffusion. Complementary inputs, in-cluding telecommunications, transport andpower infrastructure, and a well-educated laborforce are critical to exploiting the Internet’s po-tential. Governments have also encouraged theexpansion of the Internet by subsidizing Inter-net connections and investing directly in infra-structure, although such investments can crowdout private initiatives and may quickly becomeobsolete due to rapid changes in technology.

Other policies can also contribute to boost-ing Internet use. An open foreign direct invest-ment regime is necessary to promote dissemi-nation of information technology and training.Governments can help facilitate services thatevaluate and attest to the quality of outputfrom domestic firms, which could supporttheir access to global Internet exchanges. Governments must provide a supportive legal framework for electronic transactions, such asrecognition of digital signatures and legal ad-missibility of electronic documentation. Gov-ernments can also encourage Internet expansionby moving procurement and administrative re-quirements (tax forms and permits, for exam-ple) online. Finally, it is desirable to avoid highlevels of taxation on critical inputs to elec-tronic commerce such as personal computersand telecommunications equipment.

The gap in Internet access between indus-trial and developing countries will persistthrough the next decade.

Access to the Internet is grossly unequal,with 30 percent of the U.S. population onlinecompared with an average of 0.6 percent in

developing countries. Access, supported bythe growing use of cell phones as a major linkto the Internet, is expected to rise at a fasterrate in developing countries than in industrialcountries during the next 10 years. Internetaccess is likely, nonetheless, to remain limitedin per capita terms, especially in the poorestcountries, and to remain well below levels al-ready achieved in industrial countries. Accessby firms in developing countries may increasesignificantly, but the poorest developing coun-tries may still see their competitiveness im-paired because of a lack of human capital andcomplementary services required for effectiveparticipation in electronic commerce.

Emergence of electronic commerce

Transacting business using electronic aidsis as old as the telegraph, which was in-

troduced in the mid-nineteenth century. Morerecently, electronic data interchange (EDI) sys-tems not residing on the Internet have facili-tated business transactions worth trillions ofdollars. This chapter, however, explores howthe relatively new phenomenon of the Internetis likely to affect commerce. Several defini-tions of electronic commerce exist, includingtransactions where the Internet is used togather information, to order goods or services,and to make payments. A reasonable definitionof electronic commerce would include com-mercial operations in which two of these threesteps are taken electronically.1

Although the chapter focuses on electroniccommerce (in keeping with the overall themeon trade), this is by no means the only, or nec-essarily the most important, way the Internetwill affect developing countries. Increased ac-cess to information holds enormous promisefor bettering noncommercial aspects of thelives of people in the developing world—pro-viding health and education services from adistance, and more efficient government ad-ministration are but two examples. At thesame time, the growth of the Internet will in-crease the exposure of developing countries tomaterial, such as pornography, that may beviewed as undesirable.

G L O B A L E C O N O M I C P R O S P E C T S

2 Embargoed until Tuesday, December 5, 2 p.m. EST

Unpublished Proofs

Electronic commerce in industrial countrieshas grown rapidly, from next to nothing in themid-1990s to $100 to $200 billion in 1999–2000 (figure 4.1). Nevertheless, the dollar valueof electronic commerce transactions is less than1 percent of the total U.S. gross domestic prod-uct (GDP) of $23 trillion (and the business-to-business data refer to total turnover, not justvalue added). Consumer Internet purchasesequal about two-thirds of 1 percent of retailsales of goods in the United States (U.S. Depart-ment of Commerce 1999), excluding servicessuch as travel, tickets, and financial brokers,and about one-third of 1 percent in the UnitedKingdom and Germany (OECD 2000a).2

The importance of electronic commercerests not in its current size but in the likelyspeed of its establishment as a significant ve-hicle for commerce and the potential for fu-ture growth. Electronic commerce is projectedto reach $4 trillion to $6 trillion in the UnitedStates alone within the next three to four years (Bermudez and others 2000; Economist1999).3 Electronic commerce may account foras much as 25 percent of world trade by 2005(UNCTAD 1999).

The digital divide

The distribution of Internet access amongcountries is severely unequal. Despite

rapid growth in Internet access in developingcountries, industrial countries still account forthe majority of Internet subscribers (figure 4.2).More than 30 percent of U.S. residents had ac-cess to the Internet in 1999, compared with0.5 percent in Sub-Saharan Africa (figure 4.3).Electronic commerce is also relatively small inmost developing countries. In Latin America,for example, electronic commerce is estimatedat $459 million in 1999 (Lapper 2000), com-pared with a GDP of about $2 trillion.

Internet access in the developing worldvaries greatly. Some countries, particularly inEast Asia, have achieved impressive penetra-tion rates. For example, the share of Internetsubscribers in Korea has grown rapidly and isestimated at 20 percent of the population in2000, above rates in most European countries(Grebb 2000). Although per capita subscriberrates in China and India remain low, thesecountries are so large that they have a criticalmass of subscribers ready to benefit from the

E L E C T R O N I C C O M M E R C E A N D D E V E L O P I N G C O U N T R I E S

3Embargoed until Tuesday, December 5, 2 p.m. EST

Unpublished Proofs

Figure 4.1 Estimates of electronic commerce in industrial countries, 1999–2000Billions of U.S. dollars

Source: See Suttle 2000 and so forth in the chapter references.

0

Suttle 2000 Teo 1999 OECD 2000 Phillips and Meeker 2000

Gartner

20

40

60

80

100

120

140

160

Business to business

Business to consumer

Business to business and consumer

Internet, a situation that increases the potentialfor electronic commerce transactions.

Analysis of Internet diffusionThe nascent condition of Internet diffusion inmany developing countries reflects the con-straints on Internet use, the most important of

which is the availability of telecommunicationsservices. Canning (1999) finds strong evidencethat the quantity and quality of telecommuni-cations services provided in a country is a sig-nificant determinant of the existence of Inter-net connections and the level of Internet use; todate, almost all Internet users have depended



Unpublished Proofs

Figure 4.2 Estimates of Internet access, 1990–2000Millions of subscribers

United States Western Europe Japan Developing countries

0

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

50

100

150

200

250

Source: Pyramid.

Figure 4.3 Regional Internet accessPercentage of population with access to the Internet

Source: Pyramid.

0Industrialcountries

UnitedStates

Other Middle Eastand

North Africa

Sub-Saharan

Africa

EasternEurope

LatinAmericaand the

Caribbean

Asia

5

10

15

20

25

30

35

on telephone lines for connection. The trendsin “Internet intensity”—the ratio of Internetsubscribers to available telephone lines—areremarkably similar across developing and in-dustrial countries, however. Urban density andthe policy environment for private sector de-velopment are strongly related to growth in In-ternet intensity.4 Many developing countries(including, on average, those in Asia, LatinAmerica, and Sub-Saharan Africa) are experi-encing much more rapid diffusion of the Inter-net for the given availability of telephone linesthan is the United States. The digital divide re-sults from differential access to telecommuni-cations, not from the use of the Internet aftertelecommunications are available.

Unfortunately, the gap in telecommunica-tions services between industrial and develop-ing countries is large, so the digital divide islikely to remain wide for some time. The gapis also wide among developing countries, withthe poorest countries being particularly disad-vantaged. For example, the average OECDcountry had 70 times, and the average LatinAmerican country had 17 times, the number oftelephone mainlines than did countries in Sub-Saharan Africa (excluding South Africa) (fig-ure 4.4). By some indicators, the digital divideis widening. Wilson and Rodriguez (2000) findthat an index of between-country inequality inaccess to communications (the components arepersonal computers, Internet hosts, fax ma-chines, mobile phones, and televisions) deteri-orated substantially during the 1990s.5

Prospects for Internet accessGiven the enormous investments required fortelephone lines (and in some countries the con-tinued dominance of the telephone system byinefficient monopolies), hopes for narrowingthe digital divide rest largely on the spread ofalternative means of accessing the Internet.The availability of cable, cellular phone, andsatellite systems is likely to reduce dependenceon telephone lines for access to the Internetduring the next decade. Digital cellular tele-phone systems with Internet access are alreadyspreading rapidly in Japan and Western Eu-

rope. Approximately 10 million users in Japan,or 40 percent of total users there, accessed theInternet through mobile telecommunicationsdevices in May 2000 (Reuters 2000). Low-Earth-orbit satellite systems also have potentialfor reaching areas where telephone service ispoor (Wood 1999).6 Work is underway to in-vestigate the feasibility of Internet transmissionthrough power lines.

All of these links to the Internet will bypassthe often inefficient and difficult-to-build tele-phone networks now used for Internet accessand will have substantially higher accessspeeds, although the forecasts in table 4.1 arespeculative. The rapid diffusion of cellularphone and other systems with Internet accesswill mean that much of the Internet’s capacity



Unpublished Proofs

Table 4.1 Future Internet access speeds

Platform Year available Potential bandwidth

Cellular 1999 144 kilobits2000 1.6 megabits2003 5.2 megabits

Cable 2000 1–10 megabitsSatellite 2005 64 megabitsPower grid (?) 2.5 gigabits (?)

Source: Harrow 2000; www.teledesic.com; www.mediafusioncorp.net.

Figure 4.4 Access to telecommunicationsPer 1,000 persons

0

Telephone mainlines Mobile phones

100

200

300

400

500

600

OECD countries

Latin Americaand the Caribbean

Sub-Saharan Africa,excluding South Africa

Source: World Bank 2000b.

may be available at relatively low cost in manydeveloping countries. Although the most re-cent experience with wireless Internet accesshas been disappointing in some respects (wit-ness the slow diffusion of Internet-enabledwireless phones in the United States), over themedium term these alternative Internet plat-forms are likely to give a significant boost tothe spread of the Internet.

Some insight into the implications of newplatforms for Internet access and electroniccommerce can be gained by looking at theprospects for diffusion of cellular telephones.During the 1990s cell phone diffusion withincountries was strongly influenced by percapita income, the change in per capita in-come, the size of the urban population, andthe strength of the policy environment facingthe private sector.7 Assuming future per capitaincome growth and policy performance will beequal to the 1990s experience, this equationforecasts very rapid growth in cell phone, andhence Internet, penetration during the nextdecade in all developing regions. Cell phoneuse in developing countries as a group wouldquadruple by 2010 compared with 1998.

Cell phone penetration would remain lowrelative to population, a projected 6 percent in 2010, compared with 2 percent in 1998.However, this figure does not imply that only6 percent of developing country residents willhave access to cell phones, given the potentialfor multiple use (although privacy concernsmay constrain multiple use in some circum-stances). For example, hundreds of people haveaccess to the single cell phone provided toeach village participating in the BangladeshVillage Pay Phone program.

Increased access to the Internet is only oneprecondition for effective participation in elec-tronic commerce. Many developing countries,particularly the poorest ones, lack the humancapital and complementary services requiredto make effective use of the latest technolo-gies. There also is concern that developingcountry firms will face increasing challenges incompeting with the leading firms in industrialcountries, which have a headstart in using these

new technologies (although at the same timedeveloping-country firms will benefit from lowerprices and increased access to services offeredby industrial country firms). Finally, an overlyrestrictive interpretation of current rules on in-tellectual property rights could constrain devel-oping countries’ access to some of these newtechnologies, which have largely been devel-oped in the industrial world. One potentialissue is the patenting of business processes andmethods linked to the Internet; for the timebeing that practice is found only in the UnitedStates. International recognition of such patentscould constrain the ability of firms in othercountries to compete.

Effects on productivity inindustrial and developingcountries

Electronic commerce will generate produc-tivity gains by reducing transaction costs.

The rapid dissemination of information, thesubstitution of digital for paper record keep-ing, and the networking capabilities of theInternet will improve flexibility and respon-siveness, encourage new and more efficient in-termediaries, increase the use of outsourcing,reduce time to market by linking orders toproduction, and improve internal coordina-tion. Although the effect of electronic com-merce on productivity has probably been smallto date (Oliner and Sichel 2000), simulationshave indicated that electronic commerce couldraise output levels by some 5 percent in themajor industrial countries (Mann, Eckert, andKnight 2000; OECD 2000a).8 Firms can ex-pect productivity gains through improved sys-tems for procurement and inventory controland reduced costs of intermediation and salestransactions, as well as through more rapiddiffusion of technology. Consumers also willbenefit through reduced search costs, thus in-creasing competition and reducing prices.

Procurement and inventory controlFirms in developing countries can use the In-ternet to achieve the kinds of procurement and



Unpublished Proofs

inventory savings now enjoyed only by thelargest firms that have established EDI sys-tems, simply by purchasing “out of the box”electronic commerce applications (box 4.1).Goldman Sachs (1999) estimates that 30 per-cent or more of the total cost of intermediategoods typically are “process costs,” or thecosts of administering transactions and main-taining inventories. The potential for savingscan be divided into reduced processing costsof procurement transactions, reduced price ofinputs attributable to increased competition,and improved inventory control.

Substituting the Internet for paper-basedsystems can reduce the cost of processing or-ders by saving staff time, speeding up theprocess, and reducing processing errors.9 Esti-mates of the savings in processing costs ofWeb-based procurement are 90–95 percent(Schwartz 2000; U.S. Department of Com-merce 1999).

The use of online auctions also can reducethe price of inputs by improving transparencyand facilitating competitive bidding. GeneralElectric, for example, has cut the cost of pur-chased inputs by 10 to 20 percent throughonline bidding. The potential savings from in-creased transparency varies with the infor-mation content of the good. Goldman Sachs(1999) estimates that the savings from pur-chasing online may vary from 2 percent inrelatively undifferentiated products (such ascoal) to 40 percent in highly differentiated ones(some electronic components, for example).

Keeping an electronic inventory and trans-ferring information on replenishment needsover the Internet enables producers and retail-ers to reduce the time that components andraw materials spend at each processing stage.Even relatively small reductions in inventoryholding time in retail trade can mean substan-tial increases in profits because the average



Unpublished Proofs

EDI systems provide one view of the potential effi-ciency gains from relying on the Internet for pro-

cessing procurement and inventory control. Thesesystems use proprietary software to connect pur-chasers’ and suppliers’ computers and to automatethe transaction processing and information ex-change. EDI is estimated to support about $3 trillionin economic activity in the United States alone(Phillips and Meeker 2000). About 80 percent of thedollar value of intercompany transactions amongFortune 500 companies in the United States is con-ducted through EDI systems (Bermudez and others2000). Despite this, only about 100,000 U.S. compa-nies use EDI—out of the 2 million U.S. companieswith 10 employees or more. The large investment re-quired to develop proprietary software and the costsinvolved in integrating new suppliers effectively barssmall firms from using EDI. By contrast, a largeshare of the investment required for similar systemsover the Internet has already been made, and interac-tivity with other computers is automatically pro-vided. Also, it is estimated that operating costs in

Box 4.1 Electronic data interchange (EDI) systemsInternet-based systems are less than 1 percent of EDIsystems (Xie 2000).

That large firms are willing to make huge fixedinvestments and pay high operating costs for EDIsystems indicates the substantial gains that firms cancapture by transferring from paper-based to elec-tronic systems. The smaller initial investment andlower operating costs in Internet-based systems(along with greater flexibility and transparency ofoperations) means that small- and medium-size firmscan capture these gains.

The migration of EDI systems to the Internet (to reduce operating costs and to improve flexibility) is likely to boost the dollar value of electroniccommerce transactions over the next few years. This change will be slow because of the reluctance to abandon the huge fixed costs represented in EDI systems and the cost of conversion to Internet-based systems (Wenninger 1999). Concerns over the lower security of the Internet compared with that of proprietary EDI systems also may limit conversions.

cost to retailers of holding inventory for a yearis at least 25 percent of the price, and marginsmay average only 3–4 percent (OECD 1999).Improved inventory control will enable firmsto become more integrated with suppliers,thereby saving time and allowing greater pro-duction specialization. Increased productionintegration has led to a boom in specializedmanufacturing firms that produce compo-nents for more well-known companies. A fa-mous example is Cisco, whose componentsare made to Cisco’s specifications by suppli-ers, tested through a connection to the Inter-net, and then shipped directly to the buyer.

Procurement in most developing countries isslower, less efficient, and more labor-intensivethan in industrial countries, so the technical ef-ficiency gains from transferring procurementsystems to the Internet could be relatively large(although the lower cost of labor in developingcountries means that the economic gains couldbe more limited than in industrial countries).The savings in working capital from reducedholding of inventories also would be significantin developing countries, where the cost ofcapital is high and credit is often rationed orunavailable. The lack of reliable telecommu-nications networks and complementary ser-vices—for example, transport facilities—maylimit these gains, however. Some limited surveyevidence (see appendix for description of sur-vey) indicates that North American firms thatwere better at supply-chain management tobegin with are cutting these costs by an evenlarger amount when using Internet-based in-ventory systems. This may be because an ade-quate supply of high-skilled workers and aflexible organization are required to reap thefull benefits of these systems.

Reduced intermediary costsProductivity gains can be derived from elimi-nating or improving the efficiency of interme-diaries involved in marketing and distribution.Middlemen often charge substantial markupsbecause of their knowledge about and con-tacts with suppliers. By greatly expanding ac-

cess to information, the Internet has enabledthe elimination of retailers, wholesalers, and(in the case of intangible products) even dis-tributors in some sectors. More commonly,existing middlemen have been replaced bynew approaches to intermediation made pos-sible by the technology—for example, onlineauctions and aggregators (firms that representcollections of buyers that can demand lowerprices for bulk purchases).

The Internet also can generate significantcost savings in transport. The advertisementand trading of empty truck space over theWeb is reducing costs per ton in the U.S.trucking sector (Economist 1999). Accordingto one industry estimate, $15 billion to$20 billion annually in cost savings (4 to5 percent of output in the U.S. trucking indus-try) may be realized (Business 2.0 1999).

Eliminating or transforming intermediaryfunctions will enable developing-country pro-ducers to access both domestic and foreignmarkets at lower cost. By contrast, firms in de-veloping countries whose main purpose is tohelp domestic companies trade with interna-tional markets will be at particular risk. Net-work externalities, combined with a low mar-ginal cost of adding new users, mean that the market for providing intermediary services offers considerable advantage to the first com-pany on the scene. Thus the later-arriving and less technologically sophisticated firms inmany developing countries may have difficultycompeting with industrial country firms as Internet-based intermediaries (UNCTAD 2000).Developing countries also may not be able tocapture the cost savings from reduced inter-mediation in some sectors, such as primarycommodity exports, where purchasers arelikely to be the major beneficiaries of any costsavings (box 4.2).

Retail transactions The Internet offers the potential for savings in retail transactions compared with tradi-tional systems. OECD (1998b) suggests thatthe greater availability of information to the



Unpublished Proofs

consumer and savings on providing servicescould increase the productivity of sales staff inOECD countries by a factor of 10. The evi-dence on the sale of goods over the Internet sofar does not show large savings, however. Pre-liminary studies found that goods sold on theInternet were priced the same or higher thanin stores (Goldman Sachs 1997; Krantz 1998;Lee, Westland, and Hong 2000; OECD 1998a).Other studies estimated that books and com-pact discs (CDs) were 10 percent cheaper onthe Internet (Economist 2000; Oliner andSichel 2000).10 The potential savings in ser-vice transactions are more impressive. For ex-ample, the total cost (including investment) ofbank transfers over the Internet is half that ofexisting automated systems and one-eighththat of transactions using tellers (WTO 1998;figure 4.5). Note that a portion of this savingsreflects efficiency gains, while another portion

reflects the transfer of costs from producers toconsumers in the form of time spent searchingthe Internet.

The lower cost of service transactions islikely to have a less significant effect in devel-oping than in industrial countries because thelower wages paid in developing countries meanthat firms have less incentive to undertake thefixed costs involved in setting up electronic sys-tems. Also, poor distribution systems, inade-quate protection against credit card fraud, andlimited consumer Internet access constrain thepotential for business to consumer commercein many developing countries.

Knowledge acquisition and technologydiffusion Easier access to knowledge through the Inter-net will speed technology diffusion, which isof critical importance to developing countries



Unpublished Proofs

Electronic commerce may affect relatively homo-geneous primary commodities less than it does

more differentiated products because most of thenecessary information is contained in the productprice. Although electronic commerce is likely to pro-vide some benefits to producers by increasing the ef-ficiency of commodity markets, the major benefitswill accrue to purchasers.

More timely access to market information aboutprices could generate benefits to producers. Small-holder farmers in remote areas could check the pricesin the nearest market (which could be a considerablejourney if done in person) before deciding whether tosell to local middlemen; such a capability could po-tentially improve the farmers’ bargaining positionwith local and foreign buyers. The Internet alsocould provide producers with better informationabout input prices and product availability and eas-ier access to training about best production practices.

Several firms have projected that a large share of commodity trade will occur over the Internet

Box 4.2 The Internet and primary commodityexporters

(for example, Forrester Research 1999a), but theefficiency gains will vary depending on the com-modity. Commodities already traded on establishedexchanges (such as wheat), with widely dissemi-nated information on prices and centralized trading,may not be greatly affected. Online auctions willhave a greater role in reducing margins for com-modities (such as fertilizer) that trade throughbrokers with limited price transparency. For exam-ple, brokerage fees in the sugar trade, which rangefrom 0.5 to 1.0 percent of the value of the com-modity, may decline as more trade is done over the Internet.

Producers in developing countries are unlikelyto see substantial increases in incomes from lowertrading and marketing costs; rather, these gains willaccrue to the consumer through lower prices for finalproducts. Lower consumer prices may increase de-mand only slightly, because the demand for mostcommodities is price inelastic and the reduction inmarketing costs will probably be small.

because they tend to operate within the tech-nological frontier. Electronic commerce canreduce the costs of communication betweengeographically distant partners and lower thesearch-and-compare costs involved in findingpotential business partners and technologies.Moreover, the Internet provides a radial struc-ture for interpersonal communication net-works. Bulletin boards and news servers allowindividuals to exchange information fasterand within a wider environment than withnetworks based on telephone and fax. Con-nolly (1998) found that differences in commu-nication and transportation infrastructurewere significantly related to differences in therate of product imitation encouraged by for-eign direct investment (although this does notnecessarily mean that electronic commerce hasan independent positive effect). Grossman andHelpman (1991) argue that international con-tacts enable a country to obtain foreign tech-nologies and adjust them to domestic use, animportant channel through which the produc-tivity levels of industrial and developing coun-tries are interrelated. Such international “net-

working” is greatly facilitated by the Internet.Harris (1998) quotes a Neilson survey thatfound business’s primary use of the Internetwas for gathering information.

Effects on international trade indeveloping countries

By opening markets to a wider range of po-tential buyers and sellers, the Internet is

likely to foster a greater volume and variety oftrade. The Internet could erode an importantadvantage now enjoyed by firms in industrialcountries: proximity to wealthy customers.For example, the Internet reduces the cost ofproducing customized products designed fordistant markets; consumers in the UnitedStates can purchase a hand-sewn suit made bya tailor in Shanghai without ever visitingChina (Xie 2000).

Service exports The Internet will reduce barriers to the sale of services embodying skilled labor (Harris1998).11 In the Philippines, for example, com-



Unpublished Proofs

Figure 4.5 Cost savings from electronic commerceRatio of cost of traditional process to Internet

Notes: Bank transfer: total cost of teller and ATM transaction versus online.Airline tickets: travel agent booking via computer reservations system versus consumer booking online.Sending documents: send a 42-page document by fax, courier, and airmail relative to e-mail.Software distribution: download to CD and ship versus electronic transmission.Source: Kibati 1999, OECD 1999, WTO 1998.

0

Bank transfer Airline tickets Sending documents Softwaredistribution

5

10

15

20

25

30

35

panies use the Internet to provide accountingservices, process insurance claims, and trackcredit card defaulters for firms in industrialcountries (Jordan and Hilsenrath 2000). In India workers have been transcribing U.S.physicians’ oral records into written files since 1996, at one-tenth the cost of U.S. tran-scription services (Mills 1996).12 Schuknechtand Perez-Esteve (1999) suggest that financial,insurance, and other business and communica-tion services are likely to see the greatest im-pact from electronic commerce.

Multinational supply chains The Internet’s impact on access to multina-tional supply chains by firms in developingcountries is uncertain. Increased informationon these firms may improve their access tomultinationals, which tend to use supplierswith whom they have experience. GoldmanSachs (1999) estimates that, because of poorresearch, firms’ purchasing managers tend toaward 90 percent of their procurement con-tracts to about 20 percent of suppliers. At thesame time, suppliers with poor hardware,software, and Internet transmission capabil-ities may be unable to compete with better-connected companies. It is unclear whetherthe new online auction systems have resultedin the expansion of supply networks. GeneralElectric, for one example, may be increasingthe number of its suppliers through its onlinebidding site for procurement.

A lack of credibility may make it difficultfor firms in developing countries to access on-line auctions. Purchasers need to have confi-dence that suppliers will provide input on timeand in conformance with specifications, andproduct quality may not be known ex ante.More than half of 35 large firms using onlineauction or exchange sites said that they wouldnot do business through online Web sites withfirms they did not know (Forrester Research1999b). Interview results indicate buyers—typically firms in industrial countries—see anespecially high risk in purchasing from firmsin developing countries. Over time, greater usemay be made of certification agencies (such as

the International Standards Organization andthe International Electrotechnical Commis-sion) to assess independently the quality ofnew firms’ products and services. However,relatively few small firms, even in industrialcountries, use the certification services thesebodies provide, because of the cost and fearsthat certification may not fully address buyers’concerns in the markets where small firmscompete (Callaghan and Schnoll 1997).

Online advertisement New websites are emerging that provide avenue for smaller firms to advertise theirgoods, and buyers to advertise their productneeds.13 A survey of one of the best-known ad-vertising websites (www.alibaba.com), whichhas grown to 200,000 subscribers since itsApril 1999 inception, indicates that the impacthas been modest so far, although it is too earlyto judge the site’s full potential. Most par-ticipants say they have found only a few newcustomers so far. Only 13 percent of firms re-ported that total sales had gone up consider-ably since posting on the website, whereas64 percent reported some increase and the restnone (no firm reported a decline in sales).However, 87 percent of the firms viewed the



Unpublished Proofs

Figure 4.6 New customers gained fromalibaba websitePercentage of respondents

0None Few Many

10

20

30

40

50

60

70

Source: See appendix.

website as helpful or very helpful, with thesmaller firms showing the greatest enthusiasm.

Effects on income distribution

The Internet improves access to and use ofinformation, which increases the produc-

tivity of capital, thus raising its return relativeto labor (Rodriguez 2000). The Internet alsoincreases the demand for skilled labor, partic-ularly in the information technology sector. Bycontrast, better information will tend to re-duce the demand for, and hence the relative re-turn to, unskilled labor involved in the routineprocessing of transactions and (perhaps) retailsales.14 There is a risk that this divergence indemand for skilled versus unskilled laborcould increase inequality between industrialand developing economies as well as withindeveloping economies.15 Any rise in inequalitymay be exacerbated by opportunities for mi-gration as skills shortages in the informationtechnology sectors intensify in industrial coun-tries. Electronic commerce also may increaseregional inequality. In many developing na-tions, Internet services rarely spread beyondthe country’s capital and a few large urbancenters. In Kenya, for example, more than

85 percent of Internet users are in Nairobi(Jensen 1999; Kibati 1999).

Some aspects of electronic commerce couldmitigate its impact on inequality, particularlyon the poor, in developing countries. The fallin production costs is likely to increase the de-mand for all workers, despite the fall in theper unit labor input in production. Althoughinequality may increase, the income of thepoor may rise. Also, electronic commerce in-creases market transparency, thus reducingsearch costs and reliance on intermediaries.These effects reduce the price of skill-intensivegoods, thus raising the real incomes of work-ers generally, although the poor are unlikely tobenefit greatly as consumers because of theirlimited access.

A direct impact of technological change oninequality in developing countries has beendifficult to show empirically, perhaps becausethe adoption of new technologies has coin-cided with economywide structural reform(Rodriguez 2000). Studies have shown thatfor industrial countries, the recent rise in in-come inequality has occurred during a periodof rapid technological change in the informa-tion technology sector.16

Impediments to Internet use andthe role of policies in developingcountries

The presence of network externalities,where all participants gain from each ad-

dition to the network, implies that marketprices may not fully reflect the total benefit to society from increased Internet access. Thusgovernment has an important role in speedingInternet diffusion. Inappropriate policies andthe lack of complementary services, particu-larly affecting the telecommunications sector,other infrastructure, human capital, and the in-vestment environment severely constrain In-ternet access in developing countries.

Telecommunications Poor telecommunications will limit the growthof electronic commerce. Required telecommu-



Unpublished Proofs

Figure 4.7 Increased sales reportedbecause of alibaba websitePercentage of respondents

0None Moderate Considerable

10

20

30

40

50

60

70

Source: See appendix.

nications facilities include transmission facili-ties connecting a country’s domestic networkto the greater Internet, the domestic Internetbackbone, and connections from homes andbusinesses to the backbone network. The de-fects of domestic telecommunications servicesmay be less important for the larger firms indeveloping countries; these firms may find itprofitable to invest in telecommunications fa-cilities (such as wireless) that bypass the localnetwork.

State or privatized monopolies that controlinternational connections impose inefficientpricing structures and conditions,17 which meansthat many Internet service providers cannotafford to buy enough transmission capacityfor electronic commerce applications to func-tion without congestion. This poor state ofdomestic backbone networks results in a largevolume of domestic Internet traffic being sentto the United States before being returned toits region of origin (Cukier 1999). A growingnumber of African Internet sites are hosted onservers in Europe or the United States becauseof poor infrastructure (Jensen 1999). Hence,

even traffic that originates and terminates do-mestically can cost the same as internationaltransmission.

The high cost of Internet access, the lack oflocal loop infrastructure necessary for basicdial-up modem access, and the poor quality ofthe local loop infrastructure that does exist allimpede connections to the domestic backbone.Country comparisons show a strong relation-ship between usage price and Internet penetra-tion; for industrial countries the correlationbetween the Internet hosts per capita and theaverage cost of Internet access from 1995 to2000 is –.73 (EU Commission 2000; OECD2000c). Developing countries face much highercosts relative to incomes than do industrialcountries (figure 4.8). For example, surveys in-dicate that some users in Beijing may spend anaverage of 35 percent of take-home salaries onInternet access charges (Rosen 1999).

Internet use appears to be higher in coun-tries where local phone service is charged at afixed rate than in those where callers are billedby the minute. For example, in most LatinAmerican countries (Mexico being the major



Unpublished Proofs

Figure 4.8 Internet monthly access charge as a percentage of GDP per capita, 1998

Source: ITU 1999.

United States

Australia

Finland

Japan

Turkey

Mexico

Senegal

Guinea

Mozambique

Ethiopia

Uganda

Sierra Leone

0 20 40 60 80 100 120 140

exception), local calls are charged per unit oftime (Oxford Analytica 1999), and telephonecharges account for 40 percent of monthly ac-cess costs (E-Marketer 2000). In the UnitedStates the marginal cost of local calls is zero.Note that subsidizing Internet access throughflat rate charges for local calls may discourageinvestment in alternative forms of Internet ac-cess that ultimately could be more efficient(EU Commission 2000).

For many developing countries, the mostimportant issue is the lack of telephone serviceto homes and businesses. Despite increases inrates of telephone line penetration during the1990s, more than one-third of the 130 devel-oping countries (excluding small islands) withdata for 1998 have fewer than 5 telephonelines per 100 inhabitants (figure 4.9). The com-parable level in the United States is 66.

The quality of access also is important, assome electronic commerce applications thatrely on sophisticated technology and high userinteractivity require low congestion and highbandwidth transmission between the user’s ac-cess device and the host server. The most pop-ular alternatives by which developing countries

can overcome inadequate local loop infrastruc-ture have been either shared facilities or wire-less local loop. Shared facilities, which involvelocal entrepreneurs selling the use of a com-puter with Internet access, are a fast and rela-tively cheap way of increasing Internet use. Forexample, the number of Internet users on eachInternet account in Egypt is estimated to be be-tween 2.5 and 4.5 people (El-Nawawy and Is-mail 1999). Public access Internet cafés exist insome 110 countries (Rao 1999).

Wireless and satellite technologies also pro-vide an alternative to the high costs and ineffi-ciencies of many domestic telecommunicationssystems. Although currently used primarily forvoice, mobile phones “soon will be a muchbetter device for many of the usual Internet ap-plications,” according to some technologists.18

Cellular phones in some developing countrieshave experienced strong growth rates and rela-tively high penetration, similar to those in in-dustrial countries. In Haiti, for example, poortelephone service (0.9 phone lines per 100people, less than half Africa’s average, andhuge waiting lists for new lines) has led to thegrowth of wireless service (Peha 1999). In1998 Ecuador, the Slovak Republic, and West-ern Samoa had ratios of cellular phone sub-scriptions to regular phone service similar tothose of industrial countries (ITU 1998). Onaverage, however, cellular phone penetrationremains well below industrial-country levels.Sub-Saharan Africa averages 5 cellular phonesper 1,000 people, compared with 265 cellularphones for every 1,000 people in high-incomecountries (World Bank 2000b).

Impressive increases in penetration can beachieved through increasing competition, al-though in some cases privatization has meantreducing subsidies to local calls, with a nega-tive impact on Web access, at least in the shortterm (Zambrano 1999). The relative level ofcapital spending on communication infra-structure and development of Internet applica-tion software generally tends to be higher andmore advanced in those industrial countriesthat liberalized telecommunications marketsearlier (OECD 2000a). Perhaps as a result of



Unpublished Proofs

Figure 4.9 Telephone mainlines per 100inhabitants, developing countries, 1998Number of countries

0Fewer than 5 5 to 20 More than 20

10

20

30

40

50

60

Note: United States = 66 telephone mainlines per 100 inhabitants.Source: World Bank data.

privatization and liberalization, Africa has re-corded its highest annual growth rate in maintelephone lines for a decade (AED 1998a).Figure 4.10 reveals the growing trend of pri-vatization in the telecommunications sectorsof developing countries.

Other infrastructure Poor infrastructure services (other than tele-communications) are an important constrainton electronic commerce. Frequent and longpower interruptions can seriously interfere withdata transmission and systems performance,so many Bangalore software firms have theirown generators (Panagariya 1999). Mail ser-vice can be unreliable, expensive, and time-consuming in many developing countries. Forexample, the unreliability of postal services inLatin America has meant that more expensivecourier services must be used to deliver goodsordered over the Internet, and in response, in-ternational courier services are setting up spe-cial distribution systems in Miami (Lapper2000; Oxford Analytica 1999).

The lack of safeguards against fraud can se-verely restrict credit card purchases, the mostcommon means of conducting transactionsover the Internet. For example, many Latin

American consumers are unwilling to pur-chase goods over the Internet because creditcard companies will not compensate holdersfor fraudulent use of cards (in many industrialcountries, cardholders have only a limited ex-posure to loss). This lack of security does notmake consumer purchases on the Internet impossible. In China, companies are depend-ing on cash payments and local distributionthrough taxis and bicycles to reach consumers(Fan 2000).

Human capitalA critical mass of highly skilled labor is neededin developing countries to supply the necessaryapplications, provide support, and disseminaterelevant technical knowledge for electroniccommerce. The work force in many develop-ing countries lacks a sufficient supply of theseskills, and the demand for this specializedlabor from industrial countries has furtherstrained the supply of this labor in developingcountries. In the mid-1990s North Americaand Europe had unfilled demand for profes-sionals trained in information technology (fig-ure 4.11). The wages of workers in informa-tion technology industries continue to risemore rapidly than those of workers in other



Unpublished Proofs

Figure 4.10 Developing country privatization in telecommunications, 1990–98Billions of U.S. dollars

0

5

10

15

20

25

30

1990 1991 1992 1993 1994 1995 1996 1997 1998

Source: World Bank 2000b.

U.S. industries (U.S. Department of Commerce1999). U.S. firms will be able to fill only half ofthe estimated 1.6 million positions open in2000 (ITAA 2000), while the shortfall of in-formation technology workers is estimated atalmost 1 million in Japan (OECD 2000b).

At the same time, Harris (1998) notes thatthe Internet also facilitates the mobility of skilledlabor services. Workers can choose to remainin their own country while exporting laborservices to higher-paying industrial countries.19

Developing countries may also reap some ben-efits from migration. For example, Indians inSilicon Valley have played a role in provid-ing capital, expertise, and business contacts to Indians in the software exporting firms ofBangalore.

Investment environment Several regulatory impediments to the wide-spread adoption of electronic commerce exist inmany developing countries. Duties and taxes oncomputer hardware and software and commu-nication equipment increase the expense of con-necting to the Internet. For example, a com-puter imported into some African countries may

be taxed at rates exceeding 50 percent. Theoverall environment for private sector activitiesis a significant determinant of Internet servicediffusion. An open foreign direct investmentregime helps promote technology diffusion,which is important to the growth of electroniccommerce. Foreign direct investment also is onechannel that could facilitate certification of do-mestic firms for access to online auctions.

Governments also can play an importantrole in supporting the certification of firms byproviding information on certification proce-dures, promoting access by domestic firms tointernational organizations and firms thatprovide certification, and perhaps subsidizingthe costs of certification to demonstrate thekinds of resources available in the domesticmarket. This role will be particularly impor-tant (at least in a transitional sense) as theintermediaries that formerly helped connectdeveloping-country firms to internationalmarkets are replaced by Web-based intermedi-aries that may have less information on devel-oping countries.

Governments must provide a supportivelegal framework for electronic transactions, in-cluding recognition of digital signatures; legaladmissibility of electronic contracts; and estab-lishment of data storage requirements in paperform, intellectual property rights for digitalcontent, liability of Internet service providers,privacy of personal data, and mechanisms forresolving disputes. The U.N. Commission onInternational Trade Law has a “Model Law onElectronic Commerce” that offers national leg-islatures legal principles and guidelines fordealing with some of these issues (Price Water-house Coopers 1999; UNCTAD 2000).

Governments also have had considerableimpact on Internet use through direct inter-ventions. Singapore is providing grants tolocal companies to encourage participation inelectronic commerce (Price Waterhouse Coop-ers 1999). Malaysia is wiring a zone south ofKuala Lumpur with fast communications(Bickers 1999). The “Wiring the Border” proj-ect is providing subsidies to small businessesalong the Mexico-U.S. border to finance Inter-



Unpublished Proofs

Figure 4.11 Information technology jobsunfilled because of skill shortages, 1998Thousands

0

UnitedStates

Other Germany Canada UnitedKingdom

50

100

150

200

250

300

350

400

Source: OECD 1998b.

net access. The U.S. Department of Defenseplayed a critical role in developing the initialnetworking technologies (Goodman and oth-ers 1994). The U.S. government also financedthe original Internet backbone until increaseddemand for services led to the creation ofcommercial backbones; a similar pattern wasfollowed in several other industrial countries(Braga and Fink 1997). Despite some successstories, however, the rapidity of technologicalchange greatly increases the riskiness of gov-ernment interventions to support Internet ac-cess. The expenditure of billions of dollars toconnect schools to the Internet through tele-phone lines could be wasted if wireless orpower line technology turns out to be less ex-pensive (Davis and Seib 2000). Furthermore,government investments may crowd out pri-vate sector initiatives that could provide ser-vices more efficiently.

Finally, government can support the spreadof the Internet by switching to online servicesfor its own transactions. Public sector procure-ment, many aspects of tax and customs ad-ministration, the processing of routine applica-tions (such as car permits and real estatelicenses) and other governmental functions can often be carried out through the Internet.Decisions on the use of the Internet in publicadministration should be based on the costs of providing services relative to paper-basedprocesses, the capability of government per-sonnel, and the extent of demand. Neverthe-less, greater government use of the Internet canplay a role in encouraging public participation.

Language That most Internet business is conducted inEnglish is currently an important constrainton using the Internet. Estimates of the share ofEnglish used on the Internet range from 70 to80 percent, but only 57 percent of Internetusers have English as their first language (ITU1999; Vehovar, Batagelj, and Lozar 1999). Percapita Internet use averages about 30 percentin those industrial countries where English iscommon, compared with about 5 percent inother industrial countries (figure 4.12). Famil-

iarity with English was the principal determi-nant of Internet use in Slovenia, for all eco-nomic groups (Vehovar, Batagelj, and Lozar1999). Conversely, Internet content is limitedin the local language of most developing coun-tries. From a commercial aspect, Schmitt(2000) found that just 37 percent of For-tune 100 Web sites support a language otherthan English.20

The amount of non-English material on theWeb is growing, however. Spanish websites inparticular are increasing, in part to serve thelarge Spanish-speaking community in theUnited States (Vogel and Druckerman 2000).Improvements in translation services (by peo-ple and machines), as well as Web browsersthat recognize characters of different lan-guages, should ease language constraints (U.S.Department of Commerce 1999). There is grow-ing recognition that English-only content is in-sufficient for an international economy.21



Unpublished Proofs

Figure 4.12 Internet use in industrialcountries by knowledge of English,1998–99Percentage of population

0English is

first languagefor many

Knowledge ofEnglish iscommon

Other

5

10

15

20

25

30

35

First group: Australia, Canada, Ireland, United Kingdom, United States.Second group: Denmark, Finland, Norway, Sweden.Third group: Austria, Belgium, France, Germany, Greece, Italy, Japan, Portugal, Spain, Switzerland.Source: www.headcount.com

Challenges to regulatory regimesin developing countries

Electronic commerce will pose difficultchallenges for government regulation of

tax and financial systems. The growth of elec-tronic commerce will encourage tax compe-tition and may facilitate some forms of taxevasion. Competitive pressures in domesticbanking systems will rise, generating substan-tial benefits to consumers and firms but po-tentially lowering the franchise value of exist-ing banks.

Tax policy The growth of electronic commerce will pre-sent some challenges to tax enforcement indeveloping countries and place increased em-phasis on improving the technological sophisti-cation of tax authorities and on internationalcoordination of tax collection efforts. The In-ternet reduces the cost to firms of being physi-cally far away from customers and increasesthe ability of companies to relocate production,because a substantial share of the work in-volved in organizing production is carried outby computers that can be located anywhere.Thus multinationals will find it easier to shiftactivities to low-tax regimes. Governments mayfind it more difficult to impose desired incometax levels on existing corporations, and com-petition among developing countries for in-vestment by multinationals may rise. This situ-ation will place a greater premium on efforts toreach agreements among developing countriesto limit this kind of competition.

In addition, some of the new transactionsconducted through electronic commerce willbe difficult to monitor. Government may notbe able to detect the transfer of digitized ma-terial and thus know that a taxable transac-tion has occurred.22 It may also be difficult tocontrol such transactions through the supplierbecause companies can easily provide suchservices from other jurisdictions. Thus effec-tive international agreements to assist with taxenforcement will be important.

The Internet increases the potential for con-sumer-to-consumer purchases and for barter

transactions, which also are difficult to moni-tor. Barter exchanges using digital money aresmall now, but the potential for growth couldbe great when a critical mass of participants is reached (Washington Post 2000). Reducedtransaction costs also might make it easier fortaxpayers to hide potentially taxable activi-ties. The impact of electronic commerce ontax evasion should not be overestimated, how-ever. The sale of domestic goods will still becontrolled by monitoring companies’ transac-tions, and imported goods will be controlledat the border.

Tax authorities will require greater exper-tise in information technology, both to im-prove the efficiency of tax administration andto enhance government ability to obtain andunderstand records of electronic transactions.In this respect, the greater transparency andease of retrieval of electronic transactions (com-pared with paper processes) could assist taxenforcement.

Financial sector and capital flows Electronic commerce could pose a significantchallenge to government regulation of the fi-nancial sector by reducing the franchise valueof domestic banks, thus increasing the incen-tive for banks to undertake excessive risk. Atthe same time, electronic commerce will gen-erate substantial benefits to consumers andfirms that rely on banking services. Competi-tion in domestic banking systems will increasebecause of reduced costs, greater access byforeign banks, and greater reliance on capitalmarkets by former bank borrowers.

Online banking may reduce banking trans-action costs by 15–25 percent compared withregular accounts (Morgan Stanley Dean Wit-ter 1999). Enhanced transparency and compe-tition will mean that a large part of these costreductions are transmitted to the consumers ofbanking services. In fact, some degree of pricecompetition among banks is already becomingapparent in the emerging economies (Gold-man Sachs 2000).

The Internet will greatly increase the abilityof foreign banks to compete in developing



Unpublished Proofs

countries’ domestic markets. The geographiclocation of the consumer and the service pro-vider will become increasingly irrelevant,eroding the local banks’ advantage of havinglarge branch networks. First-mover advan-tages, economies of scale, and reputationaladvantages backed by strong supervisory sys-tems will strengthen the competitive positionof industrial-country banks. They will be in abetter position than domestic banks to offerintegrated trade payments systems, in whichcustomers can obtain applications for a letterof credit, care of credit approvals, telegraphictransfers, invoices, and confirmations (Granit-sas 2000). One indication of the potential forforeign firms’ inroads into developing countrybanking systems comes from a 1997 study,which indicated that U.S. and U.K. firms dom-inated global Internet financial services mar-ket because of their reputations, head start,and the predominance of English on the Inter-net (OECD 1999).

Furthermore, the Internet will reduce theadvantage that domestic banks enjoy fromhaving a monopoly on information abouttheir clients. Local banks often are betterplaced than foreign lenders to monitor the fi-nancial position of domestic firms (Eichen-green and Mody 1999). This advantage hasenabled domestic banks to play an importantrole in on lending international capital flowsto domestic borrowers; domestic banks ac-counted for as much as 46 percent of net long-term capital flows to developing countries in1997 (World Bank 1999). Domestic banks mayexperience sharper competition for makingloans in their own markets, as the Internetmakes information about borrowers more ac-cessible to international lenders.

The easier dissemination of credit informa-tion made possible by the Internet is likely tostrengthen the trend toward greater relianceon bond financing in developing countries,potentially at the expense of domestic banks.The number of firms in developing countrieswith access to international bond markets in-creased more than sixfold from 1991 to 1998(Eichengreen and Mody 1999), and the share

of private source debt that was held in bondsin developing countries rose from about one-fourth in 1990 to more than one-half in 1999(figure 4.13). Currently, leading rating com-panies rate only several hundred corporateand foreign bonds. A private initiative is un-derway to rate millions of companies on theInternet by merging credit management tech-niques with collecting and organizing infor-mation on companies (UNCTAD 2000; www.cface.com). Furthermore, with a huge pool ofrated companies, it would become possible toissue guarantees based on estimated defaultprobabilities and to securitize these guaranteesthrough the capital markets.

International coordination Electronic commerce raises several regulatoryissues that should be addressed through im-proved international coordination. The im-portance of international cooperation to taxadministration was noted earlier. Ensuringopen access for the international transmissionof goods delivered electronically would facili-tate the continued growth of electronic com-merce. Members of the World Trade Organi-zation have decided provisionally to exemptsuch goods from customs duties.23 A more im-



Unpublished Proofs

Figure 4.13 Bond versus bank financingMillions of U.S. dollars

1,200

1,000

800

600

400

200

0

1980 1990 1999

Commercial banks

Bonds

Source: World Bank 2000.

portant contribution to ensuring open accesswould be to secure agreement on barrier-freetreatment of electronically delivered services,by strengthening commitments entered intounder the General Agreement on Trade in Ser-vices (Mattoo and Schuknecht 2000).

More generally, the rise in the volume ofcross-border transactions promised throughthe spread of the Internet will raise a host ofissues requiring international agreement. Ex-amples include improved procedures for allo-cating domain names used for Internet loca-tions;24 agreement on privacy standards fordata generated through commercial transac-tions; closer coordination of antitrust reviewsto reduce the administrative burden imposedon companies; and efforts to harmonize thelaws governing electronic commerce transac-tions. Electronic commerce will further speedthe process of globalization and thus will un-derline the importance of an effective interna-tional framework for cross-border transactions.

Notes1. See OECD 1999 and UNCTAD 2000 for a

more detailed discussion of the definition of electroniccommerce.

2. These estimates exclude sales where the Inter-net is used as an information resource only. In someservice sectors, the penetration of electronic commerceis relatively high. In the United States, for example,electronic commerce accounts for some 25 percent ofstock trades.

3. These figures vary greatly depending on the de-finition of electronic commerce used. Projections oftrillions of dollars of electronic commerce in part re-flect the migration of EDI systems to the Internet.Moreover, these forecasts are extremely speculative. Asindicated earlier, estimates of even the current size ofelectronic commerce differ substantially.

4. The regression is log(I/T 1997) – log(I/T 1990)= –14.76 – 1.04log(I/T 1999) + .86log(urban pop1999) – .10log(income 1990) + .61log(policy). All co-efficients are significant except income. Regionaldummy variables are also included. R-squared is .97.I/T stands for Internet intensity.

5. Between-country inequality of access to com-ponents of this index (such as mobile phones) appearsto have declined during the 1990s. Thus trends in thedigital divide may differ significantly among differentkinds of communications media.

6. Companies such as Teledesic and Skybridgeplan to launch satellite systems capable of supportinghigh-speed Internet access for millions of users. So farthe failure of the Iridium system has not derailed theseplans.

7. The regression equation is log(cell 99) – log(cell90) = –9.16 – .84log(cell 90) + .78log(urban pop 90) +2.12[log(income 99) – log(income 90)] + .78log(in-come 90) + 1.06log(policy). R-squared is .78. Obser-vations = 99.

8. These studies are by their nature speculative,however, and the results require several restrictive as-sumptions that may or may not reflect actual events(OECD 2000a).

9. At Cisco the replacement of phone, fax, and e-mail ordering by electronic commerce systems cut thenumber of orders that had to be reworked from 25 per-cent to 2 percent (OECD 1999).

10. The use of discounts to increase market share,coupled with the lack of profits, leaves some doubt asto whether these lower prices reflect efficiency gains,however.

11. See World Bank 1995 for an earlier discussionof the potential for long-distance service exports fromdeveloping countries.

12. The potential for exports of Internet-basedservices by developing countries in the medical sector isstaggering. About one-third of the cost of health carein the United States, or some $300 billion a year, rep-resents the cost of capturing, storing, and processinginformation such as records, physicians’ notes, test re-sults, and insurance claims (Evans and Wurster 1997).

13. Examples include www.indiaonestop.com;www.in-business.com.ar/mall/;www.maquilamarket.com; and mm.malaysiadirectory.com/b2b/, which focusmainly on markets in India, Argentina, Mexico, andMalaysia, respectively. The World Bank is providingsupport to the Virtual Souk, an online marketplace forartisans in the Middle East and North Africa, owned bylocal nongovernmental organizations and cooperatives.

14. Note that the Internet also enables servicetransactions that would not otherwise have occurred,which increases the demand for literate but not highlyskilled workers in developing countries.

15. Inequality is affected by many factors, and thestrength of this effect is unknown. Thus, whether theInternet will ultimately have a significant impact on in-equality remains uncertain.

16. For example, see Autor, Katz, and Krueger1998; Bresnahan and Brynjolfsson 1998; Krueger 1993.

17. In Egypt, for example, El-Nawawy and Ismail(1999) report that the cost of an international half cir-cuit can be 2.5 times the international tariff.

18. Helft 2000.19. Harris (1998) goes further to say that “the

Internet can eliminate the scale disadvantage of small



Unpublished Proofs

regions in producing services . . . [and] then can po-tentially lead to in-migration of skilled labor to theregion.”

20. DePalma, McCarthy, and Armstrong (1998)found that although 49 of 50 companies surveyed hadoperations outside the United States, and 80 percentprint marketing material in other languages, only fivesaid that their multilingual sites were as functionallyrich as their English sites.

21. Advertisements have appeared in U.S. busi-ness magazines highlighting the need for Internet content to be written in other languages in addition to English. Schmitt (2000) warns businesses that English-only sites are no longer feasible for interna-tional companies.

22. According to Bach and Erber (1999), it is vir-tually impossible to enforce taxes on electronic trans-actions. This conclusion is uncertain, however, and itsaccuracy will depend on whether the evolution of In-ternet technology ultimately favors privacy over gov-ernment monitoring.

23. The impact of this decision on governmentrevenues should be slight because the tariff lost if notaxes are levied on digitizable goods is less than one-fifth of 1 percent of total revenues in the major devel-oping country importers (Matoo and Schuknecht2000).

24. The World Intellectual Property Organizationissued a detailed report on the intellectual propertyissues associated with domain names and has devel-oped an online system to assist in resolving disputes(UNCTAD 2000).

ReferencesAED (Africa Economic Digest). 1998a. “African Tele-

coms: A Renaissance.” 19 (10).———. 1998b. “Roadbuilding on the Information Su-

perhighway.” 19 (10).Autor, Katz, and Krueger. 1998. “Computing Inequal-

ity: Have Computers Changed the Labour Mar-ket?” Quarterly Journal of Economics 113(4):1,169–1,213 (November).

Bach, Stefan, and Georg Erber. 1999. “Opportunitiesand Risks of Global Electronic Business Transac-tions.” Economic Bulletin. German Institute forEconomic Research. Vol. 36, no. 4. April.

Bermudez, John, Bob Kraus, David O’Brien, BobParker, and Larry Lapide. 2000. B2B CommerceForecast: $5.7T by 2004. AMR Research (http://www.amrresearch.com).

Bickers, Charles. 1999. “Asia’s Race to Go Digital.”Far East Asian Economic Review. July 1. 2.

Braga, Carlos A., and Carten Fink. 1997. “The PrivateSector and the Internet.” Public Policy for the Pri-vate Sector. Note No. 122. World Bank, Wash-ington, D.C. July.

Bresnahan and Brynjolfsson. 1998. “Information Tech-nology, Workplace Organisation and the Demandfor Skilled Labour: Firm-level Evidence.” Stan-ford University, Palo Alto, Calif. Processed.

Business 2.0. 1999. “Wise Load.”Callaghan, Nancy, and Leo Schnoll. 1997. “ISO 9000

for Small Companies.” Quality Digest Online. Canning, David. 1999. “Infrastructure’s Contribution

to Aggregate Output.” Policy Research WorkingPaper 2246. World Bank, Washington, D.C.

Connolly, Michelle. 1998. “The Dual Nature of Trade:Measuring Its Impact on Imitation and Growth.”Duke University Working Paper 97–34 (http://www.econ.duke.edu/Papers/Abstracts97/abstract.97.34.html).

Cukier, Kenneth. 1999. “Bandwidth Colonialism? TheImplications of Internet Infrastructure on Interna-tional E-Commerce.” (Paper delivered at the an-nual INET conference sponsored by the InternetSociety, San Jose, Calif., (http://www.isoc.org/inet99/proceedings/1e/1e_2.htm).

Davis, Bob, and Gerald Seib. 2000. “Technology WillTest a Washington Culture Born in IndustrialAge.” Wall Street Journal. May 1.

DePalma, D., J. McCarthy, and A. Armstrong. 1998.“Strategies for Global Sites.” The Forrester Re-port 3 (3). May.

Economist. 1999. “A Survey of Business and the Inter-net.” June 16-July 2. Vol. 351, no. 8125, B1–B39.

Eichengreen, Barry, and Ashoka Mody. 1999. “Lend-ing Booms, Reserves, and the Sustainability ofShort-Term Debt: Inferences from the Pricing of Syndicated Banking Loans.” Working Paper7113. National Bureau of Economic Research,Cambridge, Mass.

El-Nawawy, Mohamed A., and Magda M. Ismail.1999. “Overcoming Deterrents and Impedimentsto Electronic Commerce in Light of Globalisa-tion: The Case of Egypt.” (Paper presented at theannual INET conference sponsored by the Inter-net Society, San Jose, Calif. (http://www.isoc.org/inet99/proceedings/1g/1g_3.htm).

E-Marketer. 2000. “Strong Net Growth South of theBorder.” March 22. (http:www.emarketer.com).

EU Commission. 2000. Europe: An Information Soci-ety for All. Progress report for Special EuropeanCouncil on Economic Reforms and Social Cohe-sion. Brussels. March 23–4.

Evans, Philip B., and Thomas S. Wurster. 1997. “Strat-egy and the New Economics of Information.”Harvard Business Review 75(5):70–82.



Unpublished Proofs

Fan, Grace. 2000. “Pedicarts Link Shanghai’s Streets tothe Internet.” New York Times. March 29.

Forrester Research. 1999a. “Online Energy Industry toReach $266 Billion by 2004.” October. Cam-bridge, Mass.

———. 1999b. “Managing e-Marketplace Risks.” De-cember. Boston, Mass.

Goldman Sachs. 1997. “Cyber Commerce: InternetTsunami.” New York. August 4.

———. 1999. “B2B: 2B or Not 2B? Version 1.1” ———. 2000. E-Finance in Asia, Part 2: Citibank–an

Asian Internet Mandarin in the Making?Goodman, S. E., L. I. Press, S. R. Ruth, and A. M.

Rutkowski. 1994. “The Global Diffusion of theInternet: Patterns and Problems.”

Granitsas, Alkman. 2000. “Tangled in the Web: TheRush into Internet Banking May Not Be Justifiedby the Number of On-Line Customers in Asia.”Far Eastern Economic Review. May 4.

Grebb, Michael. 2000. “Korea’s Digital Quest.” Busi-ness Week. September 25.

Grossman, Gene M., and Elhanan Helpman. 1991. In-novation and Growth in the Global Economy.Cambridge, Mass. MIT Press.

Harris, R. G. 1998. “The Internet as a GPT: FactorMarket Implications.,” In General Purpose Tech-nologies and Economic Growth, edited by El-hanan Helpman. Cambridge, Mass.: MIT Press.

Harrow, Jeffrey. 2000. “The Rapidly Changing Face of Computing” (http://www.compaq.com/rcfoc/[April 24]).

Helft, D. 2000. “Latin America Looking to Wireless.”The Standard. 9 May 2000.

ITAA (Information Technology Association of America).2000. Bridging the Gap: Information TechnologySkills for a New Millennium. Arlington, Va.

ITU (International Telecommunications Union). 1998.ITU Telecommunication Indicators Handbook.Geneva.

———. 1999. Challenges to the Network—Internetfor Development. Geneva. February

Jensen, Mike. 1999. African Internet Status.Jordan, Miriam and Jon E. Hilsenrath. 2000. “Amer-

ica Talks, India Types Up the Transcript.” TheWall Street Journal. March 16.

Kibati, Mugo. 1999. “What Is the Optimal Technologi-cal and Investment Path to ‘Universal’ WirelessLocal Loop Deployment in Developing Coun-tries.” (Paper presented at the annual INET con-ference sponsored by the Internet Society, San Jose,Calif. (http://www.isoc.org/inet99/proceedings/1c/1c_2.htm).

Krantz, Michael. 1998. “The Internet Economy.” Time152 (July 20): pp. 34–41.

Krueger, Anne O. 1993. “How Computers HaveChanged the Wage Structure: Evidence from Mi-

crodata, 1984–1989.” Quarterly Journal of Eco-nomics 108 (1): 33–60. February.

Lapper, R. 2000. “Tropical Nets.” Financial Times.February 3.

Lee, Ho Guen, J. Christopher Westland, and SewonHong. 1999–2000. “The Impact of ElectronicMarketplaces on Product Prices: An EmpiricalStudy of AUCUNET.” International Journal ofElectronic Commerce. Winter 4(2): 45.

Mann, Catherine L., Sue E. Eckert, and Sarah CleelandKnight. 2000. Global Electronic Commerce: APolicy Primer. Washington, D.C.: Institute for In-ternational Economics.

Matoo, Aaditya, and Ludger Schuknecht. 2000. TradePolicies for Electronic Commerce. World Bank,Washington, D.C. Processed.

Morgan Stanley Dean Witter. 1999. “The Internet andFinancial Services.” New York. August. EquityResearch, North America.

OECD (Organisation for Economic Co-operation andDevelopment). 1998a. “Electronic Commerce:Prices and Consumer Issues for Three Products:Books, Compact Discs, and Software.” Paris.

———. 1998b. The Economic and Social Impact ofElectronic Commerce: Preliminary Findings andResearch Agenda. Paris. October.

———. 1999. The Economic and Social Impact ofElectronic Commerce: Preliminary Findings andResearch Agenda. Paris.

———. 2000a. “E-Commerce: Impacts and PolicyChallenges.” Economics Department WorkingPaper 252. Paris.

______ 2000b. Information Technology Outlook.Paris.

______ 2000c. “Local Access Pricing and E-Com-merce.” Directorate for Science, Technology andIndustry, Paris.

Oliner, Stephen D., and Daniel E. Sichel. 2000. “TheResurgence of Growth in the Late 1990s: Is In-formation Technology the Story?” Finance andEconomics Division. Series 2000–20. Federal Re-serve Board, Washington, D.C.

Oxford Analytica. 1999. Latin America: ElectronicCommerce. September 16. http://www.oxan.com.

Panagariya, Arvind. 1999. “E-Commerce, WTO, andDeveloping Countries.” UNCTAD. Geneva. July(http://www.unctad.org/en/docs/ecwto/pdf).

Peha, Jon M. 1999. “Alternative Paths to Internet In-frastructure: The Case of Haiti.” Paper preparedfor the annual INET conference sponsored by theInternet Society, San Jose, Calif. (http://www.isoc.org/inet99/proceedings/3f/3f.2htm)

Phillips, Charles, and Mary Meeker. 2000. The B2BInternet Report. Morgan Stanley Dean Witter Eq-uity Research. April. New York.



Unpublished Proofs

Price Waterhouse Coopers. 1999. SME ElectronicCommerce Study. Final report, Asia Pacific Eco-nomic Cooperation Telecommunications Work-ing Group.

Rao, MaDanmohan. 1999. “Bringing the Net to theMasses: Cybercafes in Latin America.” On theInternet. 5 (1). (http://www.indialine.com/net.editorial/editorial/70.html.

Reuters. 2000. “It’s a Wireless World in Japan.” June20.

Rodriguez, F. C. 2000. “Does Information TechnologyRaise Inequality?” University of Maryland, De-partment of Economics. College Park, Md.

Rosen, Daniel H. 1999. “Hype versus Hope for E-Commerce in China.” The China Business Re-view. July/August. 26(4): 38–42.

Schmitt, E. 2000. “The Multilingual Site Blueprint.”The Forrester Report. June 2000 (http://www.forrester.com).

Schuknecht, L., and Perez-Esteve, R. 1999. “A Quanti-tative Assessment of Electronic Commerce.”WTO Working Paper ERAD 99–01. World TradeOrganization, Economic Research and AnalysisDivision. Geneva. September.

Schwartz, Nelson. D. 2000. “Playing the Internet’sNext Gold Rush.” Fortune. May 15: 141 (10)178–82.

Suttle, Philip. 2000. “The Digital Economy and theGlobal Economy.” World Financial Markets.Morgan Guaranty Trust Company Economic Re-search. January 14.

Teo, Pebble. 1999. “Business to Business ElectronicCommerce: the Asian Experience.” Presentationto OECD Workshop. Oslo, June 17.

UNCTAD (United Nations Conference on Trade andDevelopment). 1999. “Report of the Pre-UNC-TAD-X Workshop on Exchange of Experiencesamong Enterprises in the Area of Electronic Com-merce. TD(X)/pc/3.” Geneva. September 14.

———. 2000. Building Confidence: Electronic Com-merce and Development. UNCTAD/SDTE/MISC.11. Geneva.

U.S. Department of Commerce. 1999. The EmergingDigital Economy II. http://www.ecommerce.gov/ede.report.html

Vehovar, Vasja, Zenel Batagelj, and Katja Lozar. 1999.“Language as a Barrier.” (Paper presented at theannual INET conference sponsored by the Inter-net Society, San Jose, Calif.

Vogel, Thomas T., and Pamela Druckerman. 2000.“Latin Internet Craze Sets Off Alarm Bells.” WallStreet Journal. February 16.

———. 1996. Washington Post.Washington Post. 2000. “The Wired Economy: A Spe-

cial Section.” April 5 issue. Washington, DC.Wenninger, John. 1999. “Business-to-Business Elec-

tronic Commerce. Current Issues in Economicsand Finance. Vol. 5 No. 10. June

Wilson, E., and Rodriguez, F. 2000. Are Poor Coun-tries Losing the Internet Revolution? cited inWorld Bank 2000.

Wood, L., 1999. “Lloyd’s Satellite Constellations.”(www.ee . surrey.ac .uk/Personal /L .Wood/constellations/overview.html). Report preparedfor the World Bank (http://www.infodev.org/library/working.htm.)

World Bank. 1995. Global Economic Prospects and theDeveloping Countries 1995. Washington. D.C.

———. 1999. Global Development Finance 1999:Analysis and Summary Tables, Washington, D.C.

———. 2000a. “The Networking Revolution. Oppor-tunities and Challenges for Developing Coun-tries”. InfoDev Working Paper. June. Global In-formation and Communications TechnologyDepartment. The World Bank.

———. 2000b. World Development Indicators 2000.Washington, D.C.

WTO (World Trade Organization). 1998. ElectronicCommerce and the Role of the WTO. SpecialStudy 2. Geneva.

Xie, Andrew. 2000. Global Economic Forum. MorganStanley Dean Witter, New York.

Zambrano, Raul. 1999. “Internet Users in Latin Amer-ica.” Presentation at UNCTAD Workshop on E-Commerce, Lima, August 4–5.



Unpublished Proofs


Firm interviews

The interview information included in thechapter is based on conversations with:

• managers or executives from five multi-national firms: Ingram Micro, The Gap,Ford, General Electric, and Infosys. Thesespecialize in computer hardware and soft-ware distribution; apparel; automobiles;electrical and lighting equipment; andcustoms software, respectively. The firstfour appeared on this year’s Fortune-200list, while the last is a major Indian-basedmultinational software company.

• representatives from: (i) the U.S.-Mexicoand US-Philippines Chambers of Com-merce; as well as from: (ii) the offices ofthe Commercial or Trade Attaches repre-senting Taiwan (China) and Brazil in theUnited States.

• participants in a conference on the May 24,2000, “Wiring the Border” program, in-cluding representatives from small-businessassociations, technology specialists, gov-ernment officials, and academic experts.

Website surveyThe conclusions on the alibaba website arebased on a survey of firms via fax, conductedin May 2000. To keep the survey manageable,the firms to be contacted were chosen from se-lected economic subsectors. Within these sub-sectors, we included all firms posting offers inthe last two weeks of April 2000—whether assellers or buyers. We received 105 completereplies to our questionnaire from the 800 firmssurveyed. A list of all respondents’ names andwebsites (if any) is attached.

Appendix:firm interviews and survey

Unpublished Proofs

Unpublished Proofs



China Yangzhou Weiteli Motor-

Manufacturing Co., Ltd.

Tianjin Printronics Circuit Corp.

Horman Company

China National Electric Wire and Cable

I/E Corp. (Xiamen Branch)

Hebei Xin Hua Li Da Sale Department

K.O.G. International Philippines., Inc.

Jiangxi Wire and Cable General Factory

Young Eak Trading Co., Ltd.

Shenzhen Tonghaisheng Investment

Development Co., Ltd.

Jiangsu I/E (Group) Corp. (Heiteng Co.,

Ltd.)

Sinochem Hebei (Shenzhen Toomly)

Import and Export Co., Ltd.

Zhejiang Yongkang Crown Power Tools

Manufacturing Co., Ltd.

Truly Sales Co.

Hongguang Electronics Import and

Export Co. (Guangzhou Office)

Chengdu Guoxin Maida Electronic Co.,

Ltd.

Catic Electronics

H2O Electronics Co.

Dong Young International Corp.

Sinoleather.com Ltd.

Chew The World

Gusung Machinery

Eros Group

Xiangshui Bearing Accessory Co., Ltd.

Shandong Gifts and Decorations Import

and Export Corp.

Dalian F.T.Z. Zhengxing International

Trading Co., Ltd.

Yuyao Kingfan Industry and Trade Co.,

Ltd.

Green Source International Group

Wellmade Industry Corp., Ltd.

Zhejiang Light Industrial Products

Import and Export Corp.

iSquare Design and Development

Shandong Xingfa Foodstuffs

Mideast Mercantile Ltd.

Xiamen Zhongxin Metal Products Co.,

Ltd.

Cintel International

Sandstone International Co., Ltd.

Shanghai Yang Ning International

Trading Co., Ltd.

Pacific Silk Route Pte.

China Shaanxi Techrun Technology

Company

Praphan Ceramica Co., Ltd.

China Tea Import and Export Corp.

Hebei Sanli Cashmere Products Co., Ltd.

Yixing Tanghan Ceramic Co., Ltd.

Xiamen Zhenhua Ind. Corp.

Ningbo Economic and Technical

Development Zone Import and Export

Corp.

Starscom Info-Tech Co., Ltd.

Seanet RS

Cixi Kaida Bearing Co., Ltd.

Kedi Hi-Tech Industrial Co., Ltd.,

Xiamen Office

Shijiazhuang Gulf Semiconductor Co.,

Ltd.

Tao’s Inc.

Feidong Foreign Economics and Trade

Corp.

China Dalian Aidi International Trade

Company

Aurora Translation Services

Cixi Fuda Bearing Co., Ltd.

Hanbit Ebenezet

Shen Zhen Xinhaowei Industrial Co.,

Ltd.

G.K. Trading Corp.

Giga Technology Co., Ltd.

On Time Taiwan Ltd.

Suzhou Arts and Crafts I/E Corp.

Belgraver Asia Pte. Ltd.

W. & J. Co., Ltd.

ChangZhou Rui Da Trading Company

Well Hung (Australia) Pte. Ltd.

Renaissance International

Wuhan Zhongbai Group Co., Ltd.

CNACC International Co., Ltd.

Fujian Coal Import and Export Corp.

Tung Kong Handbag Mfy

Ambp Enterprises Co., Ltd.

Software International

Xiamen Gemachieve Enterprise

Shandong Metals and Minerals Import

and Export Corp.

D.P. International

Jitco Group Ltd.

Beijing Orient Sotoma Garment Co., Ltd.

Ishida Co., Ltd.

Ningbo Free Trade Zone Sino-Dubai

International Trading Co., Ltd.

Shriya Impex

Regan (H. K.) Ltd.

Citic Shanghai Import and Export Co.,

Ltd.

Chengde Bearing Co., Ltd.

Manray Concept

Atul Exports

W. H. Enterprises

Guangdong Yangchun Bearing Co. Ltd.

S.L.S. Partnership

Adore International

Dorly International Enterprises Inc.

Shandong Commercial Group

Corporation

Goldsense Technology Ltd.

CV RJR International

Companies Participating in Survey of Alibaba B2B Website Users

Electronic Commerce and Developing...

Documents

Transcript of Electronic Commerce and Developing...